Law Review
Northwestern Law
Northwestern University Law Review : Colloquy : 2010 : Rethinking the Order of Battle in Constitutional Torts: A Reply to John Jeffries

Rethinking the Order of Battle in Constitutional Torts: A Reply to John Jeffries

Nancy Leong[*]

[download pdf]


The Supreme Court’s decision in Pearson v. Callahan ended an eight-year experiment in the adjudication of qualified immunity claims.[1]  That experiment began with Saucier v. Katz, in which the Court held that lower courts must decide whether a government officer violated a plaintiff’s constitutional rights before addressing the question of whether the government officer was entitled to immunity.[2]  The Court’s rationale for requiring lower courts to first address the merits was the need to clarify constitutional law for the benefit of both government actors (who could then better conform their behavior to constitutional standards) and future plaintiffs (who could then overcome the defense of qualified immunity and recover damages for their injuries).[3]  But Pearson overturned Saucier’s mandate, holding that merits-first adjudication, while often appropriate, "should no longer be regarded as mandatory."[4]  The Court cited a number of reasons for its decision, including the detriment to judicial efficiency; the reality that principles articulated may be of little value, particularly if a higher court is about to pass on the same question or if the question involves state law; the difficulty of making law on uncertain facts when qualified immunity is asserted at the pleading stage; the concern that the government will be unable to appeal an unfavorable decision on the merits if it prevails on qualified immunity; and contravention of the longstanding principle of constitutional avoidance.[5]

I have previously argued that Saucier’s merits-first approach was misguided.[6]  I based my conclusion both on various analytic criticisms of mandatory merits-first adjudication[7] and on an original empirical study indicating that the merits-first approach led disproportionately to the articulation of law-narrowing constitutional rights.[8]  In explaining the latter, I hypothesized that the close relationship between the merits and qualified immunity inquiries engenders cognitive disincentives for judges to recognize a constitutional violation, yet grant qualified immunity.[9]  Unsurprisingly, I agree with the result in Pearson; my chief complaint is that the Court provided insufficient guidance to lower courts as to when they should decide the constitutional question.[10]

In a recent article, John Jeffries critiqued Pearson while engaging recent scholarship on qualified immunity, including my own.[11]  He expressed skepticism that the qualified immunity inquiry engenders cognitive disincentives to reach certain results.[12]  He also contended that Pearson failed to adequately analyze the competing interests embedded in the question of merits adjudication.[13]  Merits avoidance entails significant costs, which, in many instances, outweigh those of merits adjudication.  First, if merits adjudication is not mandatory, judges may fail to resolve important constitutional questions.[14]  And second, in areas of the law where constitutional tort actions are the primary vehicle for articulating constitutional principles, merits-avoidance will effectively narrow the scope of rights by rendering it perpetually indefinite.[15]  As he has argued elsewhere with respect to remedies,[16] Jeffries advocated "disaggregating" our approach to lawmaking in section 1983 and Bivens actions by evaluating whether constitutional tort actions are the primary means of remedying violations in a particular doctrinal area.[17]  If they are, he believes, then merits adjudication in qualified immunity actions is both desirable and essential.[18]

Jeffries’ article presents many valuable insights.  I concur both with his characterization, also developed elsewhere, of qualified immunity as one way of removing disincentives for innovation[19] and with his ongoing advocacy of an approach to constitutional torts that takes account of available alternative remedies.[20]  I write this brief Reply in the hope that clarifying the points on which we do disagree will advance the broader discussion.

We part company on two grounds.  First, Jeffries’ view is that clarification of constitutional law is inherently desirable.[21]  While I do not, of course, advocate confusion in constitutional interpretation, I remain unconvinced that any law is always better than no law.  That is, if law articulated in certain circumstances is ill-considered or prematurely issued, I do not view the existence of that law as a benefit in itself.  I am particularly concerned with situations in which factors extrinsic to the substantive merits influence courts’ articulation of legal principles.  As my previous research suggests, factors particular to the context of qualified immunity adjudication raise concerns that the law articulated in that context will tend to narrow constitutional rights.[22]

This concern leads to another of greater generality.  Jeffries’ priority is for the law to be clarified.[23]  In advancing this proposition, he assumes that one forum for law articulation is as good as another—for example, a suppression hearing in a criminal prosecution is functionally identical to a damages action under section 1983.  Yet critical differences among the various contexts where constitutional lawmaking occurs translate to substantive differences in the resulting law.  Our decisions about where law is articulated are decisions with substantive consequences.  I argue, therefore, that we should consider the characteristics of a particular context when we are deciding whether we want constitutional rights to be clarified in that area.


Is it good for courts to articulate constitutional law?  Some would argue that clarity in the law is always better.  As the Supreme Court noted in Saucier, law articulation benefits defendants in constitutional tort suits, who then know how to conform their conduct to the law, as well as plaintiffs, who then can recover for violations of their constitutional rights.[24]  Yet, as scholars have also asserted, uncertainty and complexity in the law are sometimes preferable.[25]

For my part, I have previously argued that courts’ clarification of constitutional questions in the context of qualified immunity is not always desirable because their decisions disproportionally result in the narrowing of plaintiffs’ constitutional rights.[26]  In my previous research, I first conducted an empirical study in which I read and coded a large random sample of cases involving qualified immunity.[27]  I found—unsurprisingly—that after Saucier, the percentage of cases in which courts articulated constitutional law increased.[28]  I also found—much more surprisingly—that the increase resulted from an increase in the percentage of rulings favoring defendants, while the percentage of rulings favoring plaintiffs remained relatively constant.[29]  Other scholars have undertaken empirical studies of the effect of Saucier and reached different conclusions,[30] although significant methodological differences render direct comparison with my work inappropriate.[31]

I focus here on the second part of my previous research and Jeffries’ discussion of it.  As one possible explanation for courts’ lopsided articulation of constitutional law, I suggested that judges experience cognitive dissonance when they acknowledge that a plaintiff suffered a violation of his constitutional rights, yet grant qualified immunity so that the injured plaintiff cannot recover.  Consequently, I theorized, judges have an incentive to avoid such dissonance by harmonizing the ruling on the constitutional merits with the ultimate denial of recovery.[32]

As Jeffries accurately observes, my suggestion is speculative.  No research has yet "purport[ed] to link cognitive dissonance with merits-first adjudication in constitutional tort cases,"[33] and I do not claim to have established, with scientific exactitude, a link between judges’ experience of cognitive dissonance and their surprising reluctance to articulate constitutional principles favorable to plaintiffs.  Rather, my project proposes a well-established theory of cognitive psychology—cognitive dissonance[34]—as one plausible explanation for the asymmetry that my empirical research revealed.  Cognitive dissonance, according to my account, need not be the only explanation, or even the most significant explanation, for judges’ rulings.  My claim is simply this: Given that judges are human, we should look to the study of human cognition as a source of possible explanations for judicial decisionmaking.

Although Jeffries questions the influence of cognitive dissonance on judicial decisionmaking, the idea that cognitive factors affect judging is by no means novel.  Indeed, Jeffries and other commentators have elsewhere acknowledged that heuristics may play a role in judicial decisionmaking.[35]

Jeffries questions my account on the ground that factors endemic to judging render cognitive dissonance inapplicable.[36]  He argues that research has demonstrated that people experience cognitive dissonance only when they make decisions freely, and judging—in Jeffries’ view—is not such a context.[37]  He argues that judges do not decide freely—that they are "constrained by the external authority of the law"—thereby dispelling any cognitive dissonance.[38]  Under this view, our judicial system itself provides an antidote to any dissonance caused by recognizing a right but denying a remedy for the violation of that right.

More broadly, Jeffries asserts that the application of cognitive dissonance theory to judging "swims upstream against the broad current of legal reasoning."[39]  Judges, he observes, routinely decide issues or claims that are related but distinct in a wide range of situations.[40]  Indeed, legal training instills this ability.  He argues, therefore, that to claim that judges are ill-equipped to mentally segregate separate issues or claims is to question their ability to perform their jobs in general, not merely in the context of qualified immunity.[41]

At a basic institutional level, Jeffries’ account of the constraints placed on judicial decisionmaking is not entirely convincing.  Most obviously, judges are appointed to life terms, and, against the background of this life tenure, their decisions are seldom reversed.[42]  Although I do not think that our judiciary is populated with rogue jurists who decide each case however they please, in practical terms, relatively little prevents such behavior.  Indeed, Jeffries’ argument that judges feel constrained by Saucier is particularly unpersuasive given that lower court judges both criticized[43] and even disregarded[44] Saucier prior to Pearson and that the Supreme Court itself decided Saucier on the basis of immunity rather than on the merits.[45]

More importantly, however, Jeffries bypasses the heart of my argument by focusing on discrediting the idea that judges operate without meaningful constraints on their decisionmaking.  My claim is not that judges operate without constraint.  Rather, my claim is that institutional and precedential factors do not completely dictate judicial decisionmaking.  Case outcomes are very seldom foregone conclusions: if the law were crystalline, rational parties would settle or drop their claims.[46]  Certainly, precedent plays a role in judges’ resolution of the outcome of particular cases, but precedent is seldom entirely dispositive, and where precedent leaves off, other factors—including cognitive dissonance—may begin to influence judicial decisionmaking.  Of course, judges may not universally experience cognitive dissonance.  The point is simply that some judges are likely influenced by cognitive dissonance some of the time, and that this influence may help to explain the empirical trends in the articulation of the law.

Indeed, judges are less likely to feel constrained by precedent in the cases with which Jeffries is most preoccupied.  The qualified immunity defense, for example, generally requires a government officer to demonstrate that the law was not clearly established, or, effectively, to show that little precedent exists.[47]  And Jeffries’ concern lies primarily with cases invoking those areas of doctrine where law clarification is most needed—cases in which, by definition, the least precedent (and therefore the least constraint on judicial decisionmaking) exists.  In such cases, judges are less likely—and, indeed, less able—to simply defer to existing precedent.

The role of cognitive factors in qualified immunity adjudications is enhanced by crucial differences between qualified immunity and other doctrines—differences that Jeffries’ critique fails to take into account.  He argues that "separate questions that must be answered separately," such as those arising in the Saucier inquiry, are doctrinally common, offering as support a non-exhaustive list of such situations: subject matter and personal jurisdiction, statutes of limitations, voluntariness of confessions, searches and seizures, and expert qualifications.[48]  The frequency with which separate questions are bundled together within the same case, he infers, presupposes that judges are able to distinguish among separate questions that are part of the same matter.[49]  Yet the mere fact that judges often decide multiple questions together does not alone rule out the possibility that such decisionmaking fosters dissonance.  Indeed, we might reasonably ask whether dissonance pervades other contexts as well, although I do not take a position on that possibility here.  More importantly, qualified immunity is significantly more likely to foster dissonance than other contexts in which separate questions are decided together.  In qualified immunity adjudications, the question of whether a constitutional violation took place overlaps considerably with the question of whether the officer who committed the alleged violation is entitled to qualified immunity.  Both questions involve a close consideration of the conduct at issue, juxtaposed with the applicable substantive law.  The same cannot be said of Jeffries’ other examples—the analysis of whether a statute of limitations has run on a claim, for example, is wholly separate from the substantive merits of the claim.  Thus, the significant overlap between the constitutional and qualified immunity questions renders the qualified immunity context uniquely likely to generate dissonance if the results reached on the two inquiries contradict one another.

Even supposing that one rejects cognitive dissonance as an influence on judicial decisionmaking, common sense suggests that an immunity result may influence a merits decision.  From a practical perspective, an opinion acknowledging a constitutional violation while granting qualified immunity is difficult to write.[50]  In the excessive force context, for example, such an opinion would involve a statement that a police officer’s use of force was unreasonable enough to be unconstitutional, but not unreasonable enough to allow the plaintiff to overcome qualified immunity.[51]  Distinguishing between these marginally different points on the continuum of reasonableness is a formidable task, one that is difficult to render with intellectual honesty.[52]  The difficulty of the task may well encourage judges to decide the claims in harmony.

Outside of the qualified immunity context, Jeffries and other commentators tend to accept that judges may be influenced by considerations aside from the constitutional merits in deciding cases, and they acknowledge that pragmatic concerns do sometimes shape substantive outcomes.[53]  Given that the influence of factors extraneous to the merits is relatively uncontroversial, it does not seem implausible that the qualified immunity determination may also exert a gravitational pull on the merits decision.[54]

None of this is to say that courts should not articulate law in qualified immunity proceedings.  Jeffries’ concern for "the underenforcement of constitutional rights while . . . uncertainty continues"[55] is one that I share.  Yet the concerns I have explored both here and in past research indicate that such articulation is not necessarily an unqualified benefit.  Asking courts to decide difficult constitutional questions in qualified immunity proceedings is strong medicine, and it may have side effects.  We should consider these side effects, and weigh them against the benefits of law articulation, as we consider whether, in a particular circumstance, it is good for courts to articulate constitutional law.


The Supreme Court’s decision in Pearson highlights the need for careful thought about law articulation.  During the past few decades, the Court’s jurisprudence has increasingly emphasized that the judiciary not only resolves disputes, but also makes law.[56]  This trend is apparent in the Court’s increasing willingness to require or permit articulation of constitutional principles even where such articulation is not strictly necessary to the outcome of a particular case.  Qualified immunity is an example of one context in which the Court permits such articulation.  Another context is harmless error, where the Court has held that courts must decide whether an error occurred at all before they determine whether that error was harmless[57]—the initial decision of whether error occurred leads to the articulation of constitutional principles en route to the decision of whether that error was harmless.  A third is the good faith exception to the exclusionary rule, where the Court has held that courts may decide whether a particular law enforcement action was in fact unconstitutional, not merely whether the officer believed it to be so.[58]  In each instance, the Court has settled on an analysis that enables the articulation of constitutional principles.

This emphasis on lawmaking invites us to think normatively about where courts should make constitutional law.  Jeffries argues, and I agree, that when we consider the process of constitutional law articulation, we should not think in isolation about suits for money damages.[59]  Instead, we should consider money-damages suits in conjunction with available alternative avenues for courts to clarify constitutional rights, such as actions for injunctive or declaratory relief, or criminal prosecutions.  As Jeffries observes, Pearson’s discussion of such alternative avenues is sorely lacking.[60]

We part company, however, with respect to how we should take these alternative avenues for lawmaking into account.  For Jeffries, the existence of an alternative avenue is the critical issue.  In the past, he has similarly advocated a "disaggregated" approach to remedying constitutional violations.[61]  Each substantive right, or doctrinal area within that right, should have its own doctrine of remedy.  Under this approach, different remedial contexts are seen as equally good.  So, for example, when it comes to providing a remedy for Fourth Amendment violations, "money damages and exclusion of evidence are substitutes,"[62] and the availability of money damages under constitutional tort actions is important if no other remedies are available.[63]

The approach Jeffries advocates for the clarification of constitutional law parallels his approach to remedying constitutional violations.  We should facilitate lawmaking in a particular context if no alternative avenues are available for law articulation.[64]  Under this view, it does not matter whether Fourth Amendment law—say, law governing the legality of a particular type of search—is made in a qualified immunity adjudication or in a ruling on the exclusion of evidence.  That lawmaking occurs somewhere is sufficient.

I question this assumption that different contexts are equivalent for law articulation purposes.  Scholars have persuasively documented the "intimate relationship" between rights and remedies—the influence that the available remedy exerts over the development of a particular right.[65]  In many instances, the law that courts make is inextricable from the remedial context in which that law is made.

In coming work, I discuss in detail the influence of context on lawmaking.[66]  For present purposes, however, I simply wish to put forward the idea that the characteristics of various contexts, including qualified immunity proceedings and suppression hearings, influence not only the outcome of individual proceedings, but also the law that results from each individual proceeding.  My previous work presents data suggesting one such contextual influence: that of a qualified immunity outcome on a merits determination.[67]  In other remedial contexts, other influences will emerge.

Consider the context of a suppression hearing, in which the proponent of the constitutional right is a criminal defendant who seeks to exclude evidence from trial.  The court’s attention is drawn to the activities of the police officers that led directly to finding the evidence.  This focus may exclude consideration of other police conduct that, if assessed in isolation, might appear problematic to a court.

One example arises in the law regulating "stops and frisks"—limited and often spontaneous police questioning and searching of citizens requires only reasonable suspicion.  In Terry v. Ohio, Officer McFadden suspected three men of planning a robbery and feared that they might be armed.[68]  He approached the men, identified himself, and asked their names.[69]  "When the men ‘mumbled something’ in response to his inquiries, [he] grabbed petitioner Terry, spun him around so that they were facing the other two [men] . . . and patted down the outside of his clothing," finding a pistol in the pocket of Terry’s overcoat.[70]

The Court’s description of the initial physical contact between Officer McFadden and Terry is vivid: the officer grabbed the suspect and spun him around, admittedly without probable cause to believe criminal activity had occurred.  Yet this use of force is wholly absent from the subsequent legal analysis, which focused solely on whether there was justification for the officer’s "limited search of the outer clothing . . . to discover weapons."[71]  As Bill Stuntz aptly notes, "stigma and police use of force, which are obviously at the heart of the Terry facts, play a surprisingly small role in Terry doctrine."[72]

The explanation for the Court’s omission is straightforward: Terry raised his claim in a suppression hearing, seeking exclusion of the pistol.[73]  The Court therefore focused on what Officer McFadden did to find the pistol—that is, his search of Terry’s pockets—rather than on the non-trivial use of force that preceded it.  If the Court had instead considered a section 1983 claim alleging that a police officer grabbed a suspect and spun him around, it would surely have analyzed the legality of those specific actions, even if it ultimately found them constitutional.  But because Terry’s facts arose in the context of a suppression hearing, the doctrine that emerged from the case disregards the use of force in determining the legality of a particular stop and frisk.

The purpose of this brief digression into criminal process is to advance a much more general proposition.  If the context of a dispute affects the substance of the law that emerges from that dispute, we should regard decisions about where constitutional law is articulated as substantive decisions.  In the aftermath of the Supreme Court’s decisions in Saucier and Pearson, dozens of scholarly works have addressed the question of whether law should be made, while scarcely any have considered how making law in one type of proceeding or another may affect what that law looks like.  Policymakers, academics, and other stakeholders think quite carefully and intentionally about the process by which legislators or other government officials craft the rules that govern behavior.  These stakeholders tend not to afford the same consideration to judicial lawmaking.  But there is no reason this should be so.  Surely the lawmaking that courts do is worthy of the same evaluation as the lawmaking of legislatures.  After all, poorly-considered law is more difficult to change than to prevent.  Like all judge-made law, it takes root and grows, eventually pollinating the larger constitutional ecosystem.

So where does this leave us with respect to merits-first adjudication of qualified immunity claims?  If, as I have argued, there exists at least a possibility that qualified immunity will skew decisions made in the context of money damages, perhaps we should consider alternative avenues through which we might encourage the articulation of constitutional law.[74]

Criminal prosecution provides one possible forum, at least for claims under the Fourth Amendment.  Such claims may be asserted defensively in suppression hearings.  Many scholars, however, have expressed concerns that the criminal context is fraught with the same problems as the qualified immunity context—that is, each situation creates incentives for the courts to define constitutional principles in a manner restrictive of individual liberty.  A judge may find it difficult to disregard the evidentiary loss in close cases—for example, if a particular search yields a large amount of contraband and the search’s legality is at least debatable, the evidentiary loss that would result from suppressing the evidence might well tip the scales in favor of finding no Fourth Amendment violation.[75]

Suits for injunctive and declaratory relief present another possibility.  Skewed judicial lawmaking seems somewhat less likely in such actions because any remedy is purely prospective and will not involve backward-looking negative consequences, such as the exclusion of evidence that would help convict a hardened criminal or the imposition of damages liability on officers who may not have known they were breaking the law.  Moreover, crafting a forward-looking principle invites judges to consider not only the parties before them, but the entire class of citizens the decision will affect.  This may reduce the possibility that a judge is influenced by factors immaterial to the merits, such as the possibly unsavory characteristics of the particular proponent of constitutional rights or the availability of qualified immunity.  But the availability of injunctive and declaratory remedies is severely limited under current standing doctrine.[76]  Although the wisdom of expanding the availability of injunctive and declaratory relief hinges on multiple considerations, we might consider the opportunity for well-considered lawmaking as one part of the calculus.

As a thought experiment, we might also imagine mechanisms for separating constitutional questions from remedial and factual influences that might result in skewed lawmaking.  We might, for instance, designate certain judges to hear the merits of cases presenting constitutional issues and to frame constitutional principles that would govern future similar cases.  In such cases, we might ask the parties to brief the merits question separately from other questions in the case, regardless of whether the case would involve qualified immunity determinations (in the civil context) or motions to exclude evidence (in the criminal context).  The judge deciding the constitutional question would thus be insulated from the potentially undesirable incentives imposed by ancillary remedial considerations.  Only if the designated judge found a constitutional violation would the case proceed to another judge to determine further issues such as qualified immunity defenses or motions to dismiss criminal charges for lack of evidence.

In future scholarship I will extend this thought experiment, with the goal of exploring more thoroughly the conditions necessary for intelligent judicial lawmaking.[77]  Here, my purpose is simply to promote evaluation of where constitutional law should be made and to encourage openness to a range of venues where clarification of constitutional principles may take place.


The articulation of constitutional rights serves many purposes.  It places both citizens and government officers on notice of the law governing their behavior.  It allows for the incremental evolution of the law in response to the development of social norms, advances in technology, and shifts in policy prerogatives.  And it allows for the continued refinement of doctrine with the benefit of experience.

But not all law clarification is created equal.  Some circumstances present a heightened risk that the law articulated will be poorly reasoned, overbroad, or simply wrong.  While actions for money damages provide an important opportunity to clarify the law, data suggests that the fact that such clarification so often takes place in the context of a qualified immunity determination renders that law susceptible to skewing.  In light of the importance of law clarification, it may sometimes still be desirable to articulate constitutional law in the context of qualified immunity adjudications.  But we should bear in mind that other venues do exist and weigh the advantages and disadvantages of those other venues in determining where, ultimately, courts should make law.


*.  Assistant Professor, William & Mary School of Law.  I am grateful for helpful comments and suggestions I received from Alan Chen, Scott Dodson, Charlotte Garden, John Jeffries, Pam Karlan, Allison Orr Larsen, Justin Marceau, and Justin Pidot.  Bill Novick provided valuable research assistance.

1.  See Morse v. Frederick, 551 U.S. 393, 432 (2007) (Breyer, J., concurring in part and dissenting in part) (stating that if it were his choice, he would "end the failed Saucier experiment now") (link).

2.  533 U.S. 194, 201 (2001) (link), overruled by Pearson v. Callahan, 129 S. Ct. 808 (2009) (link).

3.  See Saucier, 553 U.S. at 200–01.

4.  Pearson, 129 S. Ct. at 818.

5.  See id. at 819–21.

6.  See Nancy Leong, The Saucier Qualified Immunity Experiment: An Empirical Analysis, 36 Pepp. L. Rev. 667, 676–84 (2009) (link).

7.  Id.

8.  Id. at 684–702.

9.  Id. at 700–08.

10.  Id. at 709.

11.  See John C. Jeffries, Jr., Reversing the Order of Battle in Constitutional Torts, 2009 Sup. Ct. Rev. 115.

12.  Id. at 124–26.  Jeffries is generally skeptical of the criticisms of mandatory merits adjudication, although he thinks they are valid in some instances, particularly with the problem of appealability.  Id. at 131.

13.  Id. at 131.

14.  See id. at 120.

15.  See id. at 131.

16.  John C. Jeffries, Jr., Disaggregating Constitutional Torts, 110 Yale L.J. 259, 290–92 (2000) (link).

17.  See Jeffries, supra note 11, at 136–37.

18.  See id.

19.  Id. at 122; John C. Jeffries, Jr., The Right-Remedy Gap in Constitutional Law, 109 Yale L.J. 87, 99–100 (1999).

20.  Jeffries, supra note 11, at 136–37; Jeffries, supra note 16, at 262–63.

21.  See Jeffries, supra note 11, at 121–26.

22.  See Leong, supra note 6, at 684–94.  I emphasize that that the narrowing of constitutional rights is not, in itself, my concern here.  Rather, my concern is that the development of constitutional law is being skewed—in this case, toward a restrictive conception of rights—by influences endemic to the qualified immunity adjudicative context.

23.  See Jeffries, supra note 11, at 131.

24.  Saucier v. Katz, 533 U.S. 194, 200–01 (2001), overruled by Pearson v. Callahan, 129 S. Ct. 808 (2009).

25.  See, e.g., Scott Dodson, The Complexity of Jurisdictional Clarity, 97 Va. L. Rev. 1 (forthcoming 2011), available at (discussing the benefits of uncertainty in jurisdictional rules) (link); Seana Valentine Schiffrin, Inducing Moral Deliberation: On the Occasional Virtues of Fog, 123 Harv. L. Rev. 1214 (2010) (describing the moral and democratic benefits that flow from vague legal standards) (link); Cass R. Sunstein, Commentary, Incompletely Theorized Agreements, 108 Harv. L. Rev. 1733, 1754–57 (1995) (defending judgments that do not fully articulate their reasoning and therefore leave the outcomes of future cases less predefined).

26.  See Leong, supra note 6, at 693.

27.  See id. at 684–88.

28.  See id. at 688–93.

29.  See id.

30.  See Paul W. Hughes, Not a Failed Experiment: Wilson-Saucier Sequencing and the Articulation of Constitutional Rights, 80 U. Colo. L. Rev. 401, 422–23 (2009) (indicating that plaintiff-friendly articulation of constitutional rights through constitutional tort suits increased significantly from the pre-Saucier time period to the post-Saucier time period); Greg Sobolski & Matt Steinberg, Note, An Empirical Analysis of Section 1983 Qualified Immunity Actions and Implications of Pearson v. Callahan, 62 Stan. L. Rev. 523, 544–48 (2010) (finding that the percentage of cases in which courts acknowledged a constitutional violation, yet granted qualified immunity, increased significantly from the pre- to post-Saucier periods) (link).

31.  I will discuss these differences in detail in future work.  Briefly, however, the Hughes and Sobolski-Steinberg studies examined only published cases.  Mine, in contrast, examined both published and unpublished cases on the grounds that judges may be intellectually influenced by evidence of their colleagues’ thoughts in previous unpublished opinions and that such opinions therefore carry informal precedential weight.  This seems particularly likely if a judge is wondering whether her opinion will be reversed on appeal, disputed by the other two members of an appellate panel, or considered for rehearing en banc.  Likewise, our studies dealt differently with cases involving multiple claims or multiple defendants.  Hughes, for instance, limited his coding to a maximum of two claims per case, while I imposed no such limitation.  In cases resolving many claims—and therefore resulting in many instances of lawmaking—limiting the number of coded claims understates the precedential impact of the case.  Finally, each of the three studies involved somewhat different time periods, which may have resulted in the variation in the results obtained.  For descriptions of these studies’ methodologies, see Hughes, supra note 30, at 418–20; Leong, supra note 6, at 684–88; Sobolski & Steinberg, supra note 30, at 539–43.

32.  Leong, supra note 6, at 702–08.

33.  Jeffries, supra note 11, at 125.

34.  Cognitive dissonance is the feeling of discomfort that results from holding two contradictory or inconsistent ideas simultaneously, which produces a drive to reduce the dissonance by modifying or rejecting one of the inconsistent ideas.  See Leon Festinger, A Theory of Cognitive Dissonance 128–29 (1957) (link).  The existence of cognitive dissonance is uncontroversial within the psychology community and is supported by volumes of research spanning more than fifty years.  See, e.g., Joel Cooper, Cognitive Dissonance: Fifty Years of a Classic Theory (2007) (link); Festinger, supra; Cognitive Dissonance: Progress on a Pivotal Theory in Social Psychology (Eddie Harmon-Jones & Judson Mills, eds., 1999) (link).

35.  For example, in discussing judicial decisions of search and seizure issues on motions to suppress evidence, Jeffries explains that, "[f]aced with a succession of guilty defendants objecting to successful searches, judges might well give officers the benefit of the doubt."  Jeffries, supra note 16, at 270.  This reasoning is essentially a version of the availability heuristic.  See Christopher Slobogin, Why Liberals Should Chuck the Exclusionary Rule, 1999 U. Ill. L. Rev. 363, 403–04 (examining the influence of representativeness and availability heuristics on the outcome of suppression hearings).

36.  See Jeffries, supra note 11, at 125.

37.  See id.

38.  Id.

39.  Id. at 126.

40.  See id.

41.  See id.

42.  Appellate judges in particular are surely well aware that the Supreme Court is highly unlikely to grant certiorari and moreover are often able to predict which cases are plausible candidates.  See, e.g., Timothy S. Bishop, Jeffrey W. Sarles & Stephen J. Kane, Tips on Petitioning for and Opposing Certiorari in the U.S. Supreme Court, 34 Litigation Magazine 26, 26–27 (2008); David C. Thompson & Melanie F. Wachtell, An Empirical Analysis of Supreme Court Certiorari Petition Procedures: The Call for Response and the Call for the Views of the Solicitor General, 16 Geo. Mason L. Rev. 237, 241 (2009) (explaining that of the 8,517 petitions filed in the Supreme Court 2005 October Term, only 78 resulted in a grant of certiorari) (link); see also Summary of the Court’s Workload, October Term 2009, (revealing that during the 2009 term the Supreme Court granted only eighty petitions for certiorari) (link).  Of course, one might argue that judges are seldom reversed precisely because they respect the constraints imposed by precedent.  While this may be true, it does not affect the reality that judges face few practical constraints on their decisionmaking.

43.  See, e.g., Dirrane v. Brookline Police Dep't, 315 F.3d 65, 69–70 (1st Cir. 2002) (explaining that Saucier's order of inquiry is "an uncomfortable exercise where . . . the answer . . . [to the constitutional question] may depend on a kaleidoscope of facts not yet fully developed" and that "[i]t may be that Saucier was not strictly intended to cover [such a] case.") (link).

44.  See, e.g., Hatfield-Bermudez v. Aldanondo-Rivera, 496 F.3d 51, 59 (1st Cir. 2007) (link); Buchanan v. Maine, 469 F.3d 158, 168 (1st Cir. 2006) (link); Santana v. Calderon, 342 F.3d 18, 29–30 (1st Cir. 2003) (link); Koch v. Town of Brattleboro, 287 F.3d 162, 168 (2d Cir. 2002) (link); see also Roberts v. Ward, 468 F.3d 963, 970 (6th Cir. 2006) (failing to segregate the constitutional rights and clearly established questions in holding that "[b]ecause the plaintiffs' allegations do not implicate any clearly established constitutional rights, we affirm the district court's grant of qualified immunity to Commissioner Ward.") (link).

45.  See Alan K. Chen, The Facts About Qualified Immunity, 55 Emory L.J. 229, 251 (2006) (link).

46.  See Frederick Schauer, Do Cases Make Bad Law?, 73 U. Chi. L. Rev. 883, 909–10 (2006).  Of course, parties are not always rational, but assuming that they generally are, a case that proceeds far enough to require a judicial ruling is more likely to hinge on unsettled law.

47.  See, e.g., Harlow v. Fitzgerald, 457 U.S. 800, 818 (1982) (link).

48.  See Jeffries, supra note 11, at 126.

49.  See id.

50.  See Diana Hassel, Excessive Reasonableness, 43 Ind. L. Rev. 117, 134–35 (2009) (describing the difficulty encountered by courts in disaggregating constitutional and qualified immunity questions) (link).

51.  See, e.g., Anderson v. Creighton, 483 U.S. 635, 643–44 (1987) (explaining that the constitutional and qualified immunity questions are distinct in the excessive force context) (link).

52.  See, e.g., Hassel, supra note 50, at 125 ("When these two standards are both operating, a court must first determine whether a defendant's actions are objectively reasonable.  Then, assuming that the actions were not objectively reasonable, the court must determine whether it was nonetheless objectively reasonable for the defendant to have believed his actions were objectively reasonable.  The application of this nonsensical series of questions leads to skewed results.").

53.  Jeffries and others agree that heuristics may affect the outcome of exclusionary proceedings.  See supra note 35.  Likewise, Jeffries and other commentators agree that absent nonretroactivity—the doctrine holding that constitutional decisions affect only plaintiffs injured in the future—courts might well avoid innovation because of the costs it would impose upon governments and institutions.  See, e.g., Jeffries, supra note 16, at 270–71.

54.  One might reasonably ask why the merits decision would not exert a countervailing gravitational pull on the immunity decision.  My data did not reveal any trend in that direction.  One possible explanation is that courts tend to consider the immediate consequences of the case first—whether an officer will face exposure to damages liability—and only then turn to the merits.

55.  Jeffries, supra note 11, at 131.

56.  See, e.g., Ashutosh Bhagwat, Separate But Equal? The Supreme Court, the Lower Federal Courts, and the Nature of the "Judicial Power", 80 B.U. L. Rev. 967, 968 (2000); Thomas Healy, The Rise of Unnecessary Constitutional Rulings, 83 N.C. L. Rev. 847, 858–71 (2005).

57.  Lockhart v. Fretwell, 506 U.S. 364, 369 n.2 (1993) ("Harmless-error analysis is triggered only after the reviewing court discovers that an error has been committed.") (link).

58.  See United States v. Leon, 468 U.S. 897, 924–25 (1984) (link).

59.  See Jeffries, supra note 11, at 131–36; Leong, supra note 6, at 709 (describing the relevance of the posture of the case and other potential contexts in which the constitutional issue could be litigated).

60.  Jeffries, supra note 11, at 131–32.

61.  Jeffries, supra note 16, at 280–82.

62.  Id. at 283.

63.  See Jeffries, supra note 11, at 117.

64.  See id. at 137; Jeffries, supra note 16, at 280–82.

65.  Daryl J. Levinson, Rights Essentialism and Remedial Equilibration, 99 Colum. L. Rev. 857, 913–14, 958 (1999); see also, e.g., Jennifer E. Laurin, Rights Translation and Remedial Disequilibration in Constitutional Criminal Procedure, 110 Colum. L. Rev. 1002, 1015 (2010) (noting that courts encounter an interpretative gap analyzing constitutional criminal rights in a civil remedial context) (link).

66.  Nancy Leong, Making Rights (November 15, 2010) (unpublished manuscript) (on file with author).

67.  See Leong, supra note 6.

68.  392 U.S. 1, 5–7 (1968) (link).

69.  Id. at 6–7.

70.  Id. at 7.

71.  Id. at 30.

72.  William J. Stuntz, Privacy’s Problem and the Law of Criminal Procedure, 93 Mich. L. Rev. 1016, 1066 (1995).

73.  Terry, 392 U.S. at 8.

74.  Indeed, the Court in Pearson suggested that these alternative avenues should affect a court’s decision whether to analyze qualified immunity in the Saucier sequence or look solely to the "clearly established" question.  See Pearson v. Callahan, 129 S. Ct. 808, 821–22 (2009).

75.  See, e.g., Akhil Reed Amar, Fourth Amendment First Principles, 107 Harv. L. Rev. 757, 799 (1994) ("Judges do not like excluding bloody knives, so they distort doctrine, claiming the Fourth Amendment was not really violated.") (link); John C. Jeffries, Jr. & George A. Rutherglen, Structural Reform Revisited, 95 Calif. L. Rev. 1387, 1407 (2007) ("Difficulties arise in borderline cases, where the mere fact that the constable blundered seems an inadequate reason to set the criminal free.  One suspects that many courts in many places strain to avoid that result." (footnote omitted)) (link); Slobogin, supra note 35, at 403 ("[R]emoving the threat of exclusion should make judges who hear Fourth Amendment claims more willing to discredit factual assertions made by the police."); George C. Thomas III & Barry S. Pollack, Saving Rights from a Remedy: A Societal View of the Fourth Amendment, 73 B.U. L. Rev. 147, 148–49 (1993) (arguing that the exclusionary remedy "encourages judges to warp Fourth Amendment doctrine and to engage in creative fact-finding").

76.  See City of Los Angeles v. Lyons, 461 U.S. 95, 105–10 (1983) (link).

77.  See Leong, supra note 66.


Copyright 2010 Northwestern University    

Cite as: 105 Nw. U. L. Rev. Colloquy 135 (2010),

Persistent URL: